I am a postdoc researcher in bioimage analysis at CBI Toulouse. Main author of Svetlana and Sketchpose Napari plugin 😎
Joined November 2023
- Tweets47
- Following86
- Followers43
- Likes48
#Sketchpose is out in @MELBAJournal !
Go read it if you are interested in training cell segmentation models using few partial annotations😀
melba-journal.org/papers/202…
Thrilled to share MyoFuse, an AI-based workflow for automated skeletal muscle cell fusion quantification! 🥳
This is a collaboration with @Benjamin_Lair_ and @CedricMoro_I2MC (@I2mcT), improving the Fusion Index (FI) measurement.
Preprint: biorxiv.org/content/10.1101/…
clement cazorla retweeted
After a long journey, Segment Anything for Microscopy is now published in Nature Methods! We significantly improve SAM for interactive and automatic segmentation in light and electron microscopy and build a user-friendly tool. nature.com/articles/s41592-0…
clement cazorla retweeted
SAMCell: Generalized Label-Free Biological Cell Segmentation with Segment Anything
Alexandra Dunnum VandeLoo, Nathan J. Malta, Emilio Aponte, Caitlin van Zyl, Danfei Xu and Craig Forest
doi.org/10.1101/2025.02.06.6…
clement cazorla retweeted
Researchers at @Imactiv3D and @maths_toulouse aim to streamline the classification of #histopathology segmentation results by minimizing the amount of manual annotation. 🖊️
Learn about the plugin they've created to address this issue in @SciReports: go.nature.com/3VCYnKE
clement cazorla retweeted
Find out more about InstanSeg from Thibaut himself at his Virtual I2K tutorial next week!
Register at i2kconference.org/virtual.ht…
Let me introduce: InstanSeg 🦠🔬💻👩🔬
This *would* have been a short thread about Thibaut Goldsborough’s PhD work… but he solved too many problems.
Now it's a long thread about 2 preprints, a whole new approach to cell segmentation & #opensource software to make it easy to use
clement cazorla retweeted
Last year, we found that diffusion models are nearly bridging the gap between human and machine (diversity vs. recognizability) for drawing arxiv.org/pdf/2301.11722.
This year, we’re exploring which inductive bias best captures human-like creativity in VAE latents 🎨
🚀🎉 Thrilled to share our #Neurips2024 paper: “Latent Representation Matters: Human-like Sketches in One-shot Drawing Tasks”! We pit humans vs. regularized Latent Diffusion Models in the one-shot drawing task. Who’s the best sketch master? 🖌️🤖 (1/5) arxiv.org/abs/2406.06079
clement cazorla retweeted
Amazing, @StRigaud is demonstrating multi-GPU support🎉 using #pyclesperanto(without _prototype) for #BioImageAnalysis at #I2K2024(virtual)!🔬🖥️🚀
The notebook is available online #openaccess
github.com/StRigaud/clespera…
clement cazorla retweeted
I just uploaded my slides for the webinar about #BioImageAnalysis Code Generation at #I2K2024 - #openaccess. See you in an hour! 🔬🖥️🤖🔓
zenodo.org/records/14001044
If you're interested in #BioImageAnalysis code generation and how to use #biabob and #gitbob, join my session at #I2K2024 virtual coming Monday 15:00 CEST and also check out the other amazing sessions! 🔬🖥️🌐🔓
i2kconference.org/workshops
clement cazorla retweeted
🚀 Thrilled to announce #inTRACKtive: a web-based tool for exploring massive cell-tracking datasets, no software installation required! Just open your browser and dive into terabytes of developmental biology data. We used it to build the virtual embryozoo.org of tracked embryonic development datasets 🐭🪰🪱🪲🐠, but you can also use it for your own data! 🔬
Preprint:
biorxiv.org/content/10.1101/…
Repository:
github.com/royerlab/inTRACKt…
This project was spear-headed by @TeunHuijben, together with engineers Andrew Sweet and @aganders3 from @cziscience
@czbbiohub #CZBiohubSF #devbio #tracking #web #visualization
🧵1/n
clement cazorla retweeted
Our new paper EDF RIM
#LSA_Highlight: [Research Article] Extended-depth of field random illumination microscopy, EDF-RIM, provides super-resolved projective imaging. @InstitutFresnel @Univ_Toulouse @univamu #Super-resolution_microscopy #Wide-field_fluorescence_microscopy
nature.com/articles/s41377-0…
Our Napari plugin Svetlana (napari-hub.org/plugins/napar…) for classifying ROIs in segmentation masks has improved. Try it!
1) The unbalance between classes is now managed better.
2) An interactive loss plotting is available in the GUI.
3) An automatic check-for-updates.
clement cazorla retweeted
I implemented an oldie-but-goodie for Jupyter Notebooks: Interactive plotting and clustering with segmented images side-by-side; new in #stackview 0.7.11: "clusterplot", inspired by the napari-clusters-plotter 🔬💻🤩
github.com/haesleinhuepf/sta…
clement cazorla retweeted
If you're not already following @Christophe21664, you definitely should! 😉
I'm super proud of this work, and there's more on the way. In case you're curious, Chris created these images without diffusion—just pure gradient ascent on a classifier. A true magician 🐇
So Im going to start being a socially engaged scientist and post papers. Here are two:
'Feature accentuation' w/ @Napoolar , Srijani Saha, @grez72, @talia_konkle - a visualization method for revealing 'what' different features 'see' in a given image.
arxiv.org/abs/2402.10039
clement cazorla retweeted
So happy to see how far @napari_imaging has gone! The repo was created with @jnuneziglesias on the Caltrain on the way to the @ComputerHistory Museum. Fast forward 6 (!) years and napari is an essential tool for bioimaging and the Python Science ecosystem. Thanks to a thriving and tireless community! Congrats to all contributors, past and present! #SoProud
6 years ago to the minute 🥳🎂, napari co-founder @loicaroyer made the 1st commit to the “napari-core” repo.
github.com/napari/napari-cor…
Since then, over 150 community members have contributed to the napari code, docs, & tooling, and over 400 plugins have been developed. #napariAt6
ALT git log output in a terminal showing the timestamp of the first commit to napari(-core)
I'm glad to announce that our paper dedicated to the classification of segmentation ROIs based on CNN has been accepted in Scientific Reports 🥳
nature.com/articles/s41598-0…
clement cazorla retweeted
A long-awaited new feature in our #BioImageAnalysis software #CellACDC is finally out!
You can now easily export to video directly from the GUI. Select your preferences (contours, timestamp, scale bar etc.), click on Export, sit back and enjoy the process 😎
Hello there !
I am now part of @batut_julie’s team at CBI (Toulouse) as a postdoctoral researcher to work on olfactive cells tracking in zebra fish embryos on 3D spinning disk microscope images !
Stay tuned 😎
cbi-toulouse.fr/fr/equipe-ba…